Authors:
Florian Baumann
1
;
Jie Lao
2
;
Arne Ehlers
1
and
Bodo Rosenhahn
1
Affiliations:
1
Leibniz Universität Hannover, Germany
;
2
USTC, China
Keyword(s):
Human Action Recognition, Volume Local Binary Patterns, Random Forest, Machine Learning, IXMAS, KTH, Weizman.
Related
Ontology
Subjects/Areas/Topics:
Applications
;
Classification
;
Computer Vision, Visualization and Computer Graphics
;
Image Understanding
;
Learning of Action Patterns
;
Object Recognition
;
Pattern Recognition
;
Software Engineering
;
Theory and Methods
Abstract:
In this paper, we propose a novel feature type to recognize human actions from video data. By combining the
benefit of Volume Local Binary Patterns and Optical Flow, a simple and efficient descriptor is constructed.
Motion Binary Patterns (MBP) are computed in spatio-temporal domain while static object appearances as
well as motion information are gathered. Histograms are used to learn a Random Forest classifier which
is applied to the task of human action recognition. The proposed framework is evaluated on the well-known,
publicly available KTH dataset, Weizman dataset and on the IXMAS dataset for multi-view action recognition.
The results demonstrate state-of-the-art accuracies in comparison to other methods.